ECP 2022 Abstract Book

Virchows Archiv (2022) 481 (Suppl 1):S1–S364 13 Methods: A total of 543 RO and 59 EoO cases were re-evaluated for basal cell hyperplasia and papillomatosis using an ocular grid and basal cell thickness (BCT) and papillary height (PH) were measured on H&E stained sections of biopsies with the most severe histology. Data were evaluated using one sample t test on SPSS version 22. A p<0.05 was considered significant. Results: EoO group (41 females, 17 males) had a significantly younger mean age of 15.25±13.9 (p<0.001) compared to RO (297 females, 246 males) with a mean age of 37.65±23.6. Papilloma- tosis was significantly more common in RO (85%) compared to EoO (55%) which showed basal cell hyperplasia (12.7% vs 100%) significantly more frequently. Morphometrically, BCT was 284 (25-250) and 153 (25-250) microns, PH was 209.75 (25-975) and 118.75 (0-325) in RO and EoO, respectively. The ratio of PH/epi- thelial thickness was significantly (p<0.05) higher (57%) in RO than EoO (39%). Although BCT was higher in RO the ratio was lower compared to EoO due to epithelial thickness which was much higher in RO. Conclusion: Morphometric analysis allowed more accurate inter- pretation of papillomatosis and basal cell hyperplasia measured as BCT and PH with respect to epithelial thickness. Indeed, pro- portioning yielded better results in distinguishing RO from EoO. The results also showed that basal cell hyperplasia proved to be a constant feature of EoO besides eosinophilia. Papillomatosis on the other hand is less discriminatory as it can be present in both conditions, though more common in RO. OFP-01-004 Clinical relevance of tumour response patterns in rectal cancer S. Kus Ozturk*, C. Graham Martínez, K. Sheahan, D. Winter, S. Aherne, E. Ryan, C.J. van de Velde, C.A. Marijnen, G.A. Hospers, R.S. van der Post, I. Nagtegaal *Radboud university medical centre, The Netherlands Background & objectives: Tumour regression grading (TRG) and tumour downstaging are standard methods for examination of pathological response after neoadjuvant chemoradiotherapy (CRT). However, different response patterns can be observed. We investigate these response patterns and their clinical relevance in rectal cancer. Methods: The study included a test and a validation cohort consisting of post-CRT rectal cancer patients with adenocarcinoma n.o.s and response to the therapy. TRG was established according to both CAP and Dworak. Response patterns were scored based on the three-step flowchart by two independent observers and correlated with pathological features and outcome. Results: The test and validation cohorts included 236 and 103 patients respectively. In both cohorts, the predominant response pattern was fragmentation (74% vs 66%) and the interobserver agreement was excellent (k=0.85). The fragmented pattern pre- sented with a significantly higher pathological stage (TNM III/ IV: 31% vs 20%; p<0.001), less tumour regression (p=0.001), a tendency towards less downstaging (downstaging: 45% vs 65%; p=0.06). The shrinkage pattern presented better overall survival (OS) (p=0.049) in the test cohort and longer recurrence-free sur- vival (RFS) (HR 2.85, 95% CI 1.09-7.49, p=0.033) in the vali- dation cohort. In the regression analysis of combined cohorts, pathological stage and Dworak TRG were independent prognostic factors of survival. Conclusion: The heterogeneous nature of tumour response fol- lowing CRT is reflected in different response patterns, which can be scored reproducibly. Fragmentation and shrinkage are the main response patterns in rectal cancer with a predominancy of the frag- mented pattern. This pattern is associated with deeper invasion and positive lymph nodes. While not independently associated with outcome, knowledge of response pattern can guide future treatment decisions, in particular concerning local treatment. OFP-01-005 Consultation rate and pathologist diagnostic rate in 6,020 oesophageal biopsy specimens R. Cheng, A. Naqvi, C. Finley, M. Bonert* *Pathology, McMaster University, Canada Background & objectives: Institution level data analyses can supplement case reviews for quality assessment. The objective of this project was to characterize the pathology in all institutional oesophageal biopsies and assess consultation rate and variation of the pathologist diagnostic rate (PDR). Methods: All in house oesophageal biopsy specimens (EBS) acces- sion 2011-2020 were extracted, diagnostically categorized with a hierarchical string-matching algorithm (HSMA) and tabulated by pathologist with documented (informal and/or formal) consulta- tions. HSMA categorizations were audited by reviewing 200 EBS reports. PDR was compared using funnel plots with 95%(p<0.05) and 99.9%(p<0.001) confidence intervals (CI) centred on the group median diagnostic rate. Results: The cohort contained 6,020 EBS and the HSMA was 99% (198/200) accurate. These were 524(8.7%) malignant/tumour, 34(0.6%) suspicious (SUSP), 72(1.2%) high-grade dysplasia (HGD), 78(1.3%) low-grade dysplasia (LGD), 65(1.1%) indefinite for dysplasia (IFD). The malignant/tumour were 339(5.6%) ade- nocarcinoma (ADN), 162(2.7%) squamous carcinoma(SCC), and 23(0.4%) other tumour(OTH). The consultation rates were 22.7%, 15.4%, 34.8%, 61.8%, 56.9%, 25.6%, 24.6% for ADN/SCC/OTH/ SUSP/HGD/LGD/IFD respectively, and 5.6% (339/6,020) for all EBS. Thirteen pathologists interpreted >150 EBS each (158-626) and together saw 5,243. The number of 95%/99.9% CI outliers were 3/1, 2/0, 1/0, 4/2, 4/4, 1/0, 5/4 for ADN/SCC/OTH/SUSP/HGD/ LGD/IFD respectively. Intestinal metaplasia (IM) (818), eosino- philic esophagitis (EE)(293) and gastroesophageal reflux (GERD) (897) had 9/4, 3/1 and 12/7 outliers respectively. Conclusion: The funnel plots demonstrated moderate to poor PDR similarity on malignant, pre-malignant and benign diagnoses. Reviews are most frequent at the benign-malignant interface. The relative rarity, and PDR variation support the recommendation for reviewing all HGD. Concordance for benign diagnoses (with the exception of EE) likely can be improved significantly. Observational data can be useful for quality assessment and for helping guide quality improvement efforts. OFP-01-006 Endoscopic biopsies in the diagnosis of colorectal carcinoma – is it quantity or quality? K. Teo*, M. Elgoweini *University Hospital Crosshouse, United Kingdom Background & objectives: Endoscopic biopsy is required not only for definitive diagnosis of colorectal carcinoma (CRC), but also for additional molecular tests. Our aim was to evaluate the relationship between the number of biopsies taken and diagnostic accuracy in our hospital. Methods: A search was performed on the departmental database for colonic biopsies obtained for suspected CRC between March and October 2021. Data was then retrieved from both endoscopy and pathology reports. Results: In total 135 cases were identified and 68% of them had ≥6 biopsies taken. Initial biopsies were reported as invasive malignancy in 101(74.8%) and the remaining 34(29.4%) reported as S2